ARCHIVES
OF BIOCHEMISTRY
ANI)
BIOPHYSICS
Vol. 292, No. 1, January, pp. 151-155, 1992
Binding of Bile Acids, Organic Anions, and Fatty Acids by Bovine lntesStinal Z Protein Hajime Takikawa,’ Department
Sakiko Arai, and Masami Yamanaka
of Medicine,
Te,ikyo University
School of Medicine, Kaga 2-11-1, Itabashi-ku,
Tokyo 113, Japan
Received July 11, 1991, and in revised form September 9, 1991
Z protein from bovine small intestinal mucosa was purified and its binding affinities for bile acids, organic anions, and fatty acids were compared with those of bovine hepatic Z protein. Purification of Z protein from intestinal and hepatic cytosol was performed by gel filtration, chromatofocusing, and hydroxyapatite chromatography. Both purified proteins had the same molecular weight (M, 14,000) and eluted from a chromatofocused gel at about pH 10. Binding studies were performed by the competitive displacement of Sanilinonaphthalene-lsulfonic acid and by equilibrium dialysis. Binding affinities for bile acids, organic anions, and fatty acids were very similar between intestinal and hepatic Z proteins. Although the real physiologic role of Z protein remains to be further elucidated, these data indicate that intestinal Z protein participates in the mechanism of intracellular bile acid tranSft?r in CnterOCyteS. 0 1992 Academic press, ho.
Bile acids play a role in enterohepatic circulation. They are effectively excreted from the liver into the bile duct and about 95% of them a.re absorbed from the intestine. Recent studies have begun to elucidate the mechanisms of bile acid transport at the brush border and the basolateral poles of enterocytes (l-4). Wilson and Treanor reported that glycocholate uptake by isolated rat ileal cells is mediated by a Nat-dependent process (1) and Weinberg et al. reported that taurocholate uptake by rat intestinal basolateral membrane vesicles is Na+ independent and insensitive to K+ potential (2). Lin et al. have identified bile acid binding polypeptides, i.e., a 99-kDa polypeptide at the brush-border membrane and both 54- and 59-kDa polypeptides at the basolateral membrane, by photoaffinity labeling of rat enterocytes (3). Recently, Simon et al. reported that the brush-,border membrane has a lower i
To
whom correspondence
slhould be addressed.
0003.9861/92 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
affinity and a higher capacity binding site for taurocholate than the basolateral membrane of rat small intestine (4). However, little is known about the mechanism of intracellular bile acid transport in enterocytes. Although a role for cytosolic proteins in this process has been suggested (5), no direct studies on the binding properties of bile acids by cytosolic proteins from the intestinal mucosa have been reported. In the liver, three families of bile acid binding proteins, Y, Y’, and Z proteins, are known and their binding properties for bile acids have been reported (6-11). Y protein has glutathione S-transferase activity (12, 13) and also binds organic anions (14-16). Y’ protein has 3cY-hydroxysteroid dehydrogenase activity in the rat liver (17) and dihydrodiol dehydrogenase activity in the human liver (18). Z protein is also known as fatty acid binding protein and has binding affinities for fatty acids and organic anions (19,20). Our recent studies suggested the physiologic importance of rat hepatic Y’ protein in intracellular bile acid transport in the liver (21, 22). In the present study, Z protein was purified to homogeneity from bovine small intestinal mucosa and its binding affinities for bile acids were compared with Z protein purified from bovine liver. Furthermore, binding affinities for organic anions and of fatty acids by intestinal and hepatic Z proteins were also studied. MATERIALS
AND
METHODS
Materials. Sephacryl S-200, Sephadex G75 superfine, PBE-118, and Pharmalyte (pH g-10.5) were purchased from Pharmacia Fine Chemicals, Inc., Lipidex 1000 was from Packard, and HA-1000 was from Tosoh. PM-10 membrane was purchased from Amicon Co. (Danvers, MA), and Spectropor 2 membrane was from Spectrum Medical Industries (Los Angeles, CA). Lithocholic and ursodeoxycholic acids were obtained from Tokyo Kasei, Inc., and other materials, including bile acids, organic anions, and fatty acids, were from Sigma. Purification of Z protein. Bovine small intestine (20 m from the terminal ileum) and bovine liver were obtained from Tokyo Zouki CO., Tokyo. Mucosa scraped from the everted small intestine and the liver were homogenized with 2 vol of 0.01 M sodium phosphate, pH 7.4, con-
151 Inc. reserved.
152
TAKIKAWA,
ARAI,
taining 0.25 M sucrose. Cytosol was obtained from the 100,OOOgsupernatant as previously reported (6, 7, 9). Gel filtration. Cytosol (100 ml) from the small intestine was eluted through a Sephacryl S-200 column (5 X 100 cm) with 0.01 M sodium phosphate buffer, pH 7.4. Azso and ANS* binding were measured in each fraction. ANS binding was determined by fluorescence spectroscopy as previously reported (6, 9). Z protein fractions from several Sephacryl S-200 columns were pooled, concentrated by PM-10 membrane, and eluted through a Sephadex G75 superfine column (5 X 100 cm) with 0.01 M sodium phosphate buffer, pH 7.4. Fractions containing Z protein corresponding to the ANS binding peak were pooled. Hepatic Z fraction was obtained by Sephadex G75 superfine chromatography of hepatic cytosol (80 ml) without Sephacryl S-200 chromatography. Intestinal and hepatic Z protein thus obtained was delipidated by elution through a Lipidex 1000 column (2.5 X 9 cm) at 37°C as previously reported (9, 23). Chromatofocusing. Delipidated intestinal and hepatic Z fractions were concentrated by PM-10 membrane and applied to a chromatofocusing gel, PBE 118 (1 X 15 cm), which had been equilibrated with 0.025 M triethylamine-HCl, pH 11. The column was eluted with a 1:45 dilution of Pharmalyte-HCl, pH 8, followed by a 1 M NaCl wash. A280, ANS binding, and pH were checked for each fraction. The major ANS binding peaks from Hydroxyapatite chromatography. chromatofocused intestinal and hepatic Z fractions were concentrated by PM-10 membrane and applied to a hydroxyapatite column (HA1000,0.74 X 7.5 cm) equipped with a Tosoh HPLC system. The column was eluted with a 0.01-0.2 M potassium phosphate gradient, pH 6.7, and ANS binding was determined in each fraction. Final fractions were subjected to SDS-PAGE according to the method of Laemmli (24) and bands were visualized by silver staining using a kit from Dai-ichi Kagaku Co., Tokyo. Varying amounts of Binding studies by ANS fEuorescence inhibition. ANS were added to cuvettes containing 2 ml of 0.01 M sodium phosphate buffer, pH 7.4, and purified Z proteins (0.066 pM for intestinal Z protein and 0.13 pM for hepatic Z protein), and the fluorescence was determined at 480 nm during excitation at 400 nm at room temperature. To elucidate the type of inhibition, ANS binding was determined with intestinal and hepatic Z protein in the presence and absence of a fixed concentration of representative inhibitors such as lithocholate, BSP, or palmitic acid. Results are expressed as Scatchard plots (25), and the dissociation constants (I&) for ANS binding were obtained by the nonlinear least-squares method using the Michaelis-Menten equation to fit the untransformed data. The inhibition constants (IQ for various ligands were determined by varying ligand concentrations at a constant ANS concentration (15 PM). K, values were calculated by the nonlinear least-squares method as previously reported (7). Binding of lithocholic and chenodeoxycholic Equilibrium dialysis. acids by purified intestinal and hepatic Z proteins was measured by equilibrium dialysis as previously reported (6-11) to confirm the results of the ANS technique. Labeled bile acids were added to 200 ~1 of protein (3.3 pM for intestinal Z protein and 6.5 j.~cM for hepatic Z protein) in Spectropor 2 membranes and dialyzed against 0.01 M sodium phosphate buffer, pH 7.4, at 4’C. After equilibrium was reached (48 h), radioactivity in the protein and buffer was determined. The Kd was approximated using a very low concentration of bile acid (Kd $ C,), where Cr is the unbound bile acid concentration, and then
Cb= -n(P).c* simplifies & +G
n(p). Cr.
to C, = -
Kd
AND
YAMANAKA
Since Cb (bound concentration), C,, P (protein concentration), (one binding site assumed) were known, Kd was estimated. RESULTS
Both intestinal and hepatic 2 proteins were purified with gel filtration, chromatofocusing, and hydroxyapatite chromatography. Figure 1 shows chromatofocusing of the intestinal Z fraction. The major ANS binding peak eluted at pH 10.2. Minor ANS binding peaks also eluted at pH 9.6 and 9.2, and after washing the column with 1 M NaCl. Figure 2 shows hydroxyapatite chromatography of the major ANS binding peak from chromatofocusing. A single ANS binding peak eluted, which was homogeneous on SDS-PAGE (M, 14,000) (Fig. 3) and used for binding studies. Figure 4 shows the chromatofocused hepatic Z fraction. The major ANS binding peak eluted at pH 10.5. Only trace ANS binding was observed in the fraction after washing with 1 M NaCl. The major ANS binding peak on chromatofocusing was further purified on hydroxyapatite chromatography (data not shown). Purified hepatic Z protein was homogeneous on SDS-polyacrylamide gel electrophoresis (iW, 14,000) (Fig. 3) and used for binding studies. Binding of bile acids, organic anions, and fatty acids by intestinal and hepatic Z protein was examined by the competitive inhibition of ANS binding. Representative
P”
0 20 Fraction
* Abbreviations used: ANS, 8-anilinonaphtalene-l-sulfonic acid; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; BSP, sulfobromophthalein; FABP, fatty acid binding proteins.
and n
40
60
number
FIG. 1. Chromatofocusing of bovine intestinal Z protein. Delipidated Z fraction (150 mg) after two sequential gel filtrations was applied to a PBE 118 gel (1 X 15 cm) equilibrated with 0.025 M triethylamine-HCl, pH 11, and eluted with a 1:45 dilution of Pharmalyte-HCl, pH 8. The flow rate was 40 ml/h and 3-ml fractions were collected.
BILE
ACID
BINDING
BY INTESTINAL
x 0.2 : =?
0.6.
0
Z PROTEIN
xxx
x Y
IM
NaCl
I
0.4.
I! 0.2.
a
I
I
\
%2:uL
0 20
10
Fraction
40 Fraction
20
30
number
FIG. 2. Hydroxyapatite chromatography of bovine intestinal 2 protein. The major ANS binding peak (Nos. 3-6, 27 mg of protein) from chromatofocusing (Fig. 1) was concentrated by PM-10 membrane and was applied to a hydroxyapatite column (0.75 X 7.5 cm) equilibrated with 0.01 M potassium phosphate bullfer, pH 6.7. The column was eluted with potassium phosphate buffer: 0.01 M for O-15 min and then a linear gradient of 0.01 to 0.1 M for 15-30 min. One-milliliter fractions were collected.
ligands competitively inhibited the binding of ANS by intestinal and hepatic Z Iproteins as shown in Scatchard plots (Fig. 5). Inhibition of the fluorescence of ANS bound at a fixed concentration to Z protein was also determined
~23
80
FIG. 4. Chromatofocusing of bovine hepatic Z protein. Delipidated Z fraction (52 mg) after Sephadex G75 superfine chromatography was applied to a PBE 118 gel (1 X 15 cm) equilibrated with 0.025 M triethylamine-HCl, pH 11, and eluted with a 1:45 dilution of PharmalyteHCl, pH 8. The flow rate was 40 ml/h and 3-ml fractions were collected. Fractions 4-9 were pooled and further purified on hydroxyapatite chromatography.
by the addition of a range of concentrations of various ligands using either intestinal (Fig. 6) or hepatic (Fig. 7) Z protein. Dissociation constants for the various ligands calculated by the nonlinear least-squares method are listed in Table I. The dissociation constants obtained by equilib-
Small
~36
60 number
Intestine
kD
-
control
-
+Lc
-
+BsP
M
+palmltac
Liver
10 2
.
acid
kD
c !k
-
-
+14kD
0
0
150
AF
FIG. 3. SDS-PAGE of purifiecd Z proteins. Chromatofocusing fractions of bovine intestinal and hepatic Z proteins were further purified by hydroxyapatite chromatography. Protein bands were visualized by silver staining.
300
0h
0
150
300
AF
FIG. 5. Scatchard plots of ANS binding by bovine intestinal (left) and hepatic (right) Z protein in the absence and presence of various inhibitors. Inhibitors included 1 @M lithocholic acid (LC), 60 PM BSP, and 8 pM palmitic acid for intestinal Z protein (0.066 PM) and 1.5 PM LC, 15 @M BSP, and 0.038 PM palmitic acid for hepatic Z protein (0.13 PM). ANS binding was determined by the change of fluorescence (Fsr = 400 nm, Fsm = 480 nm) in 2 ml of 0.01 M phosphate buffer, pH 7.4, at room temperature.
154
TAKIKAWA,
100 CA
ARAI,
TCA
UDC
50 r, OO
400
800 PM
FIG. 6. Inhibition of the fluorescence of ANS (15 PM) bound to bovine intestinal Z protein (0.066 pM) by various ligands. The fluorescence was determined at 480 nm during excitation at 400 nm in 2 ml of 0.01 M sodium phosphate buffer, pH 7.4, at room temperature. Abbreviations: LC, lithocholic acid; TLC, taurolithocholic acid; CDC, chenodeoxycholic acid, TCDC, taurochenodeoxycholic acid; DC, deoxycholic acid, TDC, taurodeoxycholic acid; UDC, ursodeoxycholic acid; CA, cholic acid; TCA, taurocholic acid, BSP, sulfobromophthalein; ICG, indocyanine green.
rium dialysis were in good agreement with the data obtained from ANS inhibition: 0.2 + 0.01 PM and 0.3 + 0.04 PM for lithocholic acid and 15 + 4 PM and 33 f 7 PM for chenodeoxycholic acid with intestinal and hepatic Z protein, respectively (means + SD of three experiments) (Table I). For comparison, the dissociation constants of various ligands by rat and human hepatic Z proteins from our previous report (9) are also included in Table I.
AND
YAMANAKA
fication steps in the present study as well as in our previous report (9), it remains unclear whether the intestinal Z protein purified in the present study corresponds to LFABP or to I-FABP. Both bovine intestinal and hepatic Z proteins have similar binding properties to rat and human hepatic Z proteins (9) with some exceptions: higher binding affinities for lithocholic acid and lower binding affinities for organic anions and oleic acid than rat and human hepatic Z proteins. The only data describing binding properties of purified intestinal Z protein are those published by Lowe et al. (29). They reported that the Kd for fatty acids coli-derived rat Z protein was 2.87-3.72 by Escherichia PM, which is similar to the values obtained from the purified bovine intestinal protein in the present study. laLin et al. (30) recently reported the photoaffinity beling of rat enterocytes using a radioactive diazo-derivative of taurocholic acid. They reported that the radioactivity after photoaffinity labeling was mostly observed in the cytosol and not in the membrane fractions. These data indicate the importance of cytosolic proteins in the intracellular transfer of bile acids in enterocytes. Lin et al. (30) also detected 14- and 35-kDa polypeptides in cytosol after photolabeling rat enterocytes. They considered that the photolabeled 14- and 35-kDa polypeptides differed from Z or Y’ protein because they were not immunoprecipitated by antibodies to Z or Y’ proteins. However, the results of the present study indicate that the photolabeled 14-kDa polypeptide in their study is Z protein itself.
DISCUSSION
We have purified bovine intestinal and hepatic Z protein using sequential gel filtration, chromatofocusing, and hydroxyapatite chromatography. Both intestinal and hepatic proteins were eluted at similar pH on chromatofocusing and have the same molecular weight. These data as well as similar binding properties for various ligands by both proteins indicate that intestinal Z protein has very similar characteristics to hepatic Z protein. In the rat, three types of fatty acid binding proteins (FABPs) have been isolated and characterized (26, 27). Among these, both intestinal (I-FABP) and liver-type (L-FABP) FABPs have been identified in the rat intestine (26,27). Although FABP has been purified from bovine liver (28), the existence of two types of FABPs in the bovine intestine is not yet known. Since we used ANS binding to monitor the identification of Z proteins in various puri-
O-M
‘-0
,uM
%
Ob
1000,uM
FIG. 7. Inhibition of the fluorescence of ANS (15 pM) bound to bovine hepatic Z protein (0.13 pM) by various ligands. The experimental conditions and abbreviations are the same as those in Fig. 6.
BILE
ACID
BINDING
BY INTESTINAL TABLE
Comparison
of Dissociation
Constants
(PM) for the Binding
Z PROTEIN
155
I
of Bile Acids,
Organic
Anions,
and Fatty
Acids by Z Proteins
Bovine Intestinal Lithocholic acid Taurolithocholic acid Chenodeoxycholic acid Taurochenodeoxycholic acid Deoxycholic acid Taurodeoxycholic acid Ursodeoxycholic acid Cholic acid Taurocholic acid Bilirubin BSP Indocyanine
green
Palmitic acid Stearic acid Oleic acid
Hepatic
0.7 (0.2) 2.8 21 (15) 350 44 350 260
Rat hepatic”
0.5 (0.3) 3.2 28 (33) 70 28 130 140
97 1700
Human hepatic”
4.8 (2.2) 7.0 71 35 6.2 25 32
160 1000
12 (9.4) 14 18 32 5.4 13 120
690
92 110
280
5.2 7.5 2.6
1.4 7.5 5.7
0.27
1.5
1.9
0.8 6.2
1.8 5.2
ND ND
ND ND
0.07
0.04
0.78 0.25
0.69 ND
Note. Dissociation constants were obtained by the ANS displacement method or direct method using equilibrium ’ Data for rat and human hepatic Z proteins are from our previous report (9).
ND
dialysis
(in parentheses).
In conclusion, Z protein may participate in the mechanism of intracellular transfer of bile acids from the brush border to the basolateral pole in enterocytes. Further investigations will be needed to examine the binding properties of the Y’ and Y proteins from the intestinal mucosa.
14. Ketley, J. N., Habig, W. H., and Jacoby, W. B. (1975) J. Biol. Chem.
REFERENCES
17. Stolz, A., Takikawa, H., Sugiyama, Y., Kuhlenkamp, witz, N. (1987) J. Clin. Inuest. 79, 427-434.
1. Wilson, F. A., and Treanor,
L. L. (1981) Gustroenterology
81, 54-
60. 2. Weinberg, S. L., Burckhardt, 3. 4. 5. 6.
G. B., and Wiloson, F. A. (1986) J. Clin. Inuest. 78, 44-50. Lin, M. C., Weinberg, S. L., Kramer, W., Burckhardt, G. B., and Wilson, F. A. (1988) J. Membr. Biol. 106, l-11. Simon, F. R., Sutherland, ,J., and Sutherland, E. (1990) Am. J. Physiol. 259, G394-G401. Wilson, F. A. (1981) Am. J. Physiol. 240, G83-G92. Sugiyama, Y., Yamada, T., and Kaplowitz, N. (1983) J. Biol. Chem.
258,3602-3607. 7. Takikawa, H., Sugiyama, Y., and Kaplowitz, N. (1986) J. Lipid Res. 27,955-966. 8. Takikawa, H., Stolz, A., Sugi.moto, M., Sugiyama, Y., and Kaplowitz, 9. 10. 11.
12. 13.
N. (1986) J. Lipid Res. 27, 652-657. Takikawa, H., and Kaplowitz, N. (1986) Arch. Biochem. Biophys. 251, 385-392. Takikawa, H., and Kaplowitz, N. (1988) J. Lipid Res. 29, 279-286. Takikawa, H., Sugiyama, Y., and Kaplowitz, J. (1988) Biochem. Biophys. Acta 954, 37-43. Kaplowitz, N., Percy-Robb, I. W., and Javitt, N. B. (1973) J. Exp. Med. 138,483-487. Habig, W. H., Pabst, M. J., Fleischner, G., Gatmaitan, Z., Arias, I. M., and Jacoby, W. B. (1974) Proc. Natl. Acad. Sci. USA 71, 3879-3882.
250,8670-8673. 15. Sugiyama, Y., Sugimoto, M., Stolz, A., and Kaplowitz, Biochem. Pharmacol. 33, 3511-3513.
N. (1984)
16. Takikawa, H., Sugiyama, Y., Stolz, A., Sugimoto, M., and Kaplowitz, N. (1986) Biochem. Pharmacol. 35, 354-356. J., and Kaplo-
18. Takikawa, H., Stolz, A., Sugiyama, Y., Yoshida, H., Yamanaka, and Kaplowitz, N. (1990) J. Biol. Chem. 265, 2132-2136. 19. Levi, A. J., Gatmaitan,
M.,
Z., and Arias, I. M. (1969) J. Clin. Znuest.
48,2156-2167. 20. Ockner, R. K., Manning,
J. A., Poppenhausen, W. K. L. (1972) Science 177, 56-58.
21. Takikawa,
H., Stolz, A., and Kaplowitz,
R. B., and Ho,
N. (1987) J. Clin. Inuest.
80,852-860. 22. Takikawa, H., Ookhtens, M., Stolz, A., and Kaplowitz, Clin. Inuest. 80, 861-866.
23. Glantz, J. F. C., Janssen, A. M., Baerwaldt,
N. (1987) J.
C. C. F., and Veerkamp,
J. H. (1983) Anal. Biochem. 837, 57-66. 24. Laemmli,
U. K. (1970) Nature 227, 680-685.
25. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 227, 680-385. 26. Bass, N. M. (1985) Chem. Phys. Lipids 38, 95-114.
27. Bass, N. M. (1988) Znt. Reu. Cytol. 3, 143-184. 28. Haunerland, N., Jagschies, G., Schulenberg, H., and Spener, F. (1984) Hoppe-Seyler’s
2. Physiol. Chem. 365, 365-376.
29. Lowe, J. B., Sacchettini, J. C., Laposata, M., McQuillan, Gordon, J. I. (1987) J. Biol. Chem. 232,5931-5937. 30. Lin, M. C., Kramer, W., and Wilson, 265, 14,986614,995.
J. J., and
F. A. (1990) J. Biol. Chem.
OF BIOCHEMISTRY
ANI)
BIOPHYSICS
Vol. 292, No. 1, January, pp. 151-155, 1992
Binding of Bile Acids, Organic Anions, and Fatty Acids by Bovine lntesStinal Z Protein Hajime Takikawa,’ Department
Sakiko Arai, and Masami Yamanaka
of Medicine,
Te,ikyo University
School of Medicine, Kaga 2-11-1, Itabashi-ku,
Tokyo 113, Japan
Received July 11, 1991, and in revised form September 9, 1991
Z protein from bovine small intestinal mucosa was purified and its binding affinities for bile acids, organic anions, and fatty acids were compared with those of bovine hepatic Z protein. Purification of Z protein from intestinal and hepatic cytosol was performed by gel filtration, chromatofocusing, and hydroxyapatite chromatography. Both purified proteins had the same molecular weight (M, 14,000) and eluted from a chromatofocused gel at about pH 10. Binding studies were performed by the competitive displacement of Sanilinonaphthalene-lsulfonic acid and by equilibrium dialysis. Binding affinities for bile acids, organic anions, and fatty acids were very similar between intestinal and hepatic Z proteins. Although the real physiologic role of Z protein remains to be further elucidated, these data indicate that intestinal Z protein participates in the mechanism of intracellular bile acid tranSft?r in CnterOCyteS. 0 1992 Academic press, ho.
Bile acids play a role in enterohepatic circulation. They are effectively excreted from the liver into the bile duct and about 95% of them a.re absorbed from the intestine. Recent studies have begun to elucidate the mechanisms of bile acid transport at the brush border and the basolateral poles of enterocytes (l-4). Wilson and Treanor reported that glycocholate uptake by isolated rat ileal cells is mediated by a Nat-dependent process (1) and Weinberg et al. reported that taurocholate uptake by rat intestinal basolateral membrane vesicles is Na+ independent and insensitive to K+ potential (2). Lin et al. have identified bile acid binding polypeptides, i.e., a 99-kDa polypeptide at the brush-border membrane and both 54- and 59-kDa polypeptides at the basolateral membrane, by photoaffinity labeling of rat enterocytes (3). Recently, Simon et al. reported that the brush-,border membrane has a lower i
To
whom correspondence
slhould be addressed.
0003.9861/92 $3.00 Copyright 0 1992 by Academic Press, All rights of reproduction in any form
affinity and a higher capacity binding site for taurocholate than the basolateral membrane of rat small intestine (4). However, little is known about the mechanism of intracellular bile acid transport in enterocytes. Although a role for cytosolic proteins in this process has been suggested (5), no direct studies on the binding properties of bile acids by cytosolic proteins from the intestinal mucosa have been reported. In the liver, three families of bile acid binding proteins, Y, Y’, and Z proteins, are known and their binding properties for bile acids have been reported (6-11). Y protein has glutathione S-transferase activity (12, 13) and also binds organic anions (14-16). Y’ protein has 3cY-hydroxysteroid dehydrogenase activity in the rat liver (17) and dihydrodiol dehydrogenase activity in the human liver (18). Z protein is also known as fatty acid binding protein and has binding affinities for fatty acids and organic anions (19,20). Our recent studies suggested the physiologic importance of rat hepatic Y’ protein in intracellular bile acid transport in the liver (21, 22). In the present study, Z protein was purified to homogeneity from bovine small intestinal mucosa and its binding affinities for bile acids were compared with Z protein purified from bovine liver. Furthermore, binding affinities for organic anions and of fatty acids by intestinal and hepatic Z proteins were also studied. MATERIALS
AND
METHODS
Materials. Sephacryl S-200, Sephadex G75 superfine, PBE-118, and Pharmalyte (pH g-10.5) were purchased from Pharmacia Fine Chemicals, Inc., Lipidex 1000 was from Packard, and HA-1000 was from Tosoh. PM-10 membrane was purchased from Amicon Co. (Danvers, MA), and Spectropor 2 membrane was from Spectrum Medical Industries (Los Angeles, CA). Lithocholic and ursodeoxycholic acids were obtained from Tokyo Kasei, Inc., and other materials, including bile acids, organic anions, and fatty acids, were from Sigma. Purification of Z protein. Bovine small intestine (20 m from the terminal ileum) and bovine liver were obtained from Tokyo Zouki CO., Tokyo. Mucosa scraped from the everted small intestine and the liver were homogenized with 2 vol of 0.01 M sodium phosphate, pH 7.4, con-
151 Inc. reserved.
152
TAKIKAWA,
ARAI,
taining 0.25 M sucrose. Cytosol was obtained from the 100,OOOgsupernatant as previously reported (6, 7, 9). Gel filtration. Cytosol (100 ml) from the small intestine was eluted through a Sephacryl S-200 column (5 X 100 cm) with 0.01 M sodium phosphate buffer, pH 7.4. Azso and ANS* binding were measured in each fraction. ANS binding was determined by fluorescence spectroscopy as previously reported (6, 9). Z protein fractions from several Sephacryl S-200 columns were pooled, concentrated by PM-10 membrane, and eluted through a Sephadex G75 superfine column (5 X 100 cm) with 0.01 M sodium phosphate buffer, pH 7.4. Fractions containing Z protein corresponding to the ANS binding peak were pooled. Hepatic Z fraction was obtained by Sephadex G75 superfine chromatography of hepatic cytosol (80 ml) without Sephacryl S-200 chromatography. Intestinal and hepatic Z protein thus obtained was delipidated by elution through a Lipidex 1000 column (2.5 X 9 cm) at 37°C as previously reported (9, 23). Chromatofocusing. Delipidated intestinal and hepatic Z fractions were concentrated by PM-10 membrane and applied to a chromatofocusing gel, PBE 118 (1 X 15 cm), which had been equilibrated with 0.025 M triethylamine-HCl, pH 11. The column was eluted with a 1:45 dilution of Pharmalyte-HCl, pH 8, followed by a 1 M NaCl wash. A280, ANS binding, and pH were checked for each fraction. The major ANS binding peaks from Hydroxyapatite chromatography. chromatofocused intestinal and hepatic Z fractions were concentrated by PM-10 membrane and applied to a hydroxyapatite column (HA1000,0.74 X 7.5 cm) equipped with a Tosoh HPLC system. The column was eluted with a 0.01-0.2 M potassium phosphate gradient, pH 6.7, and ANS binding was determined in each fraction. Final fractions were subjected to SDS-PAGE according to the method of Laemmli (24) and bands were visualized by silver staining using a kit from Dai-ichi Kagaku Co., Tokyo. Varying amounts of Binding studies by ANS fEuorescence inhibition. ANS were added to cuvettes containing 2 ml of 0.01 M sodium phosphate buffer, pH 7.4, and purified Z proteins (0.066 pM for intestinal Z protein and 0.13 pM for hepatic Z protein), and the fluorescence was determined at 480 nm during excitation at 400 nm at room temperature. To elucidate the type of inhibition, ANS binding was determined with intestinal and hepatic Z protein in the presence and absence of a fixed concentration of representative inhibitors such as lithocholate, BSP, or palmitic acid. Results are expressed as Scatchard plots (25), and the dissociation constants (I&) for ANS binding were obtained by the nonlinear least-squares method using the Michaelis-Menten equation to fit the untransformed data. The inhibition constants (IQ for various ligands were determined by varying ligand concentrations at a constant ANS concentration (15 PM). K, values were calculated by the nonlinear least-squares method as previously reported (7). Binding of lithocholic and chenodeoxycholic Equilibrium dialysis. acids by purified intestinal and hepatic Z proteins was measured by equilibrium dialysis as previously reported (6-11) to confirm the results of the ANS technique. Labeled bile acids were added to 200 ~1 of protein (3.3 pM for intestinal Z protein and 6.5 j.~cM for hepatic Z protein) in Spectropor 2 membranes and dialyzed against 0.01 M sodium phosphate buffer, pH 7.4, at 4’C. After equilibrium was reached (48 h), radioactivity in the protein and buffer was determined. The Kd was approximated using a very low concentration of bile acid (Kd $ C,), where Cr is the unbound bile acid concentration, and then
Cb= -n(P).c* simplifies & +G
n(p). Cr.
to C, = -
Kd
AND
YAMANAKA
Since Cb (bound concentration), C,, P (protein concentration), (one binding site assumed) were known, Kd was estimated. RESULTS
Both intestinal and hepatic 2 proteins were purified with gel filtration, chromatofocusing, and hydroxyapatite chromatography. Figure 1 shows chromatofocusing of the intestinal Z fraction. The major ANS binding peak eluted at pH 10.2. Minor ANS binding peaks also eluted at pH 9.6 and 9.2, and after washing the column with 1 M NaCl. Figure 2 shows hydroxyapatite chromatography of the major ANS binding peak from chromatofocusing. A single ANS binding peak eluted, which was homogeneous on SDS-PAGE (M, 14,000) (Fig. 3) and used for binding studies. Figure 4 shows the chromatofocused hepatic Z fraction. The major ANS binding peak eluted at pH 10.5. Only trace ANS binding was observed in the fraction after washing with 1 M NaCl. The major ANS binding peak on chromatofocusing was further purified on hydroxyapatite chromatography (data not shown). Purified hepatic Z protein was homogeneous on SDS-polyacrylamide gel electrophoresis (iW, 14,000) (Fig. 3) and used for binding studies. Binding of bile acids, organic anions, and fatty acids by intestinal and hepatic Z protein was examined by the competitive inhibition of ANS binding. Representative
P”
0 20 Fraction
* Abbreviations used: ANS, 8-anilinonaphtalene-l-sulfonic acid; SDS, sodium dodecyl sulfate; PAGE, polyacrylamide gel electrophoresis; BSP, sulfobromophthalein; FABP, fatty acid binding proteins.
and n
40
60
number
FIG. 1. Chromatofocusing of bovine intestinal Z protein. Delipidated Z fraction (150 mg) after two sequential gel filtrations was applied to a PBE 118 gel (1 X 15 cm) equilibrated with 0.025 M triethylamine-HCl, pH 11, and eluted with a 1:45 dilution of Pharmalyte-HCl, pH 8. The flow rate was 40 ml/h and 3-ml fractions were collected.
BILE
ACID
BINDING
BY INTESTINAL
x 0.2 : =?
0.6.
0
Z PROTEIN
xxx
x Y
IM
NaCl
I
0.4.
I! 0.2.
a
I
I
\
%2:uL
0 20
10
Fraction
40 Fraction
20
30
number
FIG. 2. Hydroxyapatite chromatography of bovine intestinal 2 protein. The major ANS binding peak (Nos. 3-6, 27 mg of protein) from chromatofocusing (Fig. 1) was concentrated by PM-10 membrane and was applied to a hydroxyapatite column (0.75 X 7.5 cm) equilibrated with 0.01 M potassium phosphate bullfer, pH 6.7. The column was eluted with potassium phosphate buffer: 0.01 M for O-15 min and then a linear gradient of 0.01 to 0.1 M for 15-30 min. One-milliliter fractions were collected.
ligands competitively inhibited the binding of ANS by intestinal and hepatic Z Iproteins as shown in Scatchard plots (Fig. 5). Inhibition of the fluorescence of ANS bound at a fixed concentration to Z protein was also determined
~23
80
FIG. 4. Chromatofocusing of bovine hepatic Z protein. Delipidated Z fraction (52 mg) after Sephadex G75 superfine chromatography was applied to a PBE 118 gel (1 X 15 cm) equilibrated with 0.025 M triethylamine-HCl, pH 11, and eluted with a 1:45 dilution of PharmalyteHCl, pH 8. The flow rate was 40 ml/h and 3-ml fractions were collected. Fractions 4-9 were pooled and further purified on hydroxyapatite chromatography.
by the addition of a range of concentrations of various ligands using either intestinal (Fig. 6) or hepatic (Fig. 7) Z protein. Dissociation constants for the various ligands calculated by the nonlinear least-squares method are listed in Table I. The dissociation constants obtained by equilib-
Small
~36
60 number
Intestine
kD
-
control
-
+Lc
-
+BsP
M
+palmltac
Liver
10 2
.
acid
kD
c !k
-
-
+14kD
0
0
150
AF
FIG. 3. SDS-PAGE of purifiecd Z proteins. Chromatofocusing fractions of bovine intestinal and hepatic Z proteins were further purified by hydroxyapatite chromatography. Protein bands were visualized by silver staining.
300
0h
0
150
300
AF
FIG. 5. Scatchard plots of ANS binding by bovine intestinal (left) and hepatic (right) Z protein in the absence and presence of various inhibitors. Inhibitors included 1 @M lithocholic acid (LC), 60 PM BSP, and 8 pM palmitic acid for intestinal Z protein (0.066 PM) and 1.5 PM LC, 15 @M BSP, and 0.038 PM palmitic acid for hepatic Z protein (0.13 PM). ANS binding was determined by the change of fluorescence (Fsr = 400 nm, Fsm = 480 nm) in 2 ml of 0.01 M phosphate buffer, pH 7.4, at room temperature.
154
TAKIKAWA,
100 CA
ARAI,
TCA
UDC
50 r, OO
400
800 PM
FIG. 6. Inhibition of the fluorescence of ANS (15 PM) bound to bovine intestinal Z protein (0.066 pM) by various ligands. The fluorescence was determined at 480 nm during excitation at 400 nm in 2 ml of 0.01 M sodium phosphate buffer, pH 7.4, at room temperature. Abbreviations: LC, lithocholic acid; TLC, taurolithocholic acid; CDC, chenodeoxycholic acid, TCDC, taurochenodeoxycholic acid; DC, deoxycholic acid, TDC, taurodeoxycholic acid; UDC, ursodeoxycholic acid; CA, cholic acid; TCA, taurocholic acid, BSP, sulfobromophthalein; ICG, indocyanine green.
rium dialysis were in good agreement with the data obtained from ANS inhibition: 0.2 + 0.01 PM and 0.3 + 0.04 PM for lithocholic acid and 15 + 4 PM and 33 f 7 PM for chenodeoxycholic acid with intestinal and hepatic Z protein, respectively (means + SD of three experiments) (Table I). For comparison, the dissociation constants of various ligands by rat and human hepatic Z proteins from our previous report (9) are also included in Table I.
AND
YAMANAKA
fication steps in the present study as well as in our previous report (9), it remains unclear whether the intestinal Z protein purified in the present study corresponds to LFABP or to I-FABP. Both bovine intestinal and hepatic Z proteins have similar binding properties to rat and human hepatic Z proteins (9) with some exceptions: higher binding affinities for lithocholic acid and lower binding affinities for organic anions and oleic acid than rat and human hepatic Z proteins. The only data describing binding properties of purified intestinal Z protein are those published by Lowe et al. (29). They reported that the Kd for fatty acids coli-derived rat Z protein was 2.87-3.72 by Escherichia PM, which is similar to the values obtained from the purified bovine intestinal protein in the present study. laLin et al. (30) recently reported the photoaffinity beling of rat enterocytes using a radioactive diazo-derivative of taurocholic acid. They reported that the radioactivity after photoaffinity labeling was mostly observed in the cytosol and not in the membrane fractions. These data indicate the importance of cytosolic proteins in the intracellular transfer of bile acids in enterocytes. Lin et al. (30) also detected 14- and 35-kDa polypeptides in cytosol after photolabeling rat enterocytes. They considered that the photolabeled 14- and 35-kDa polypeptides differed from Z or Y’ protein because they were not immunoprecipitated by antibodies to Z or Y’ proteins. However, the results of the present study indicate that the photolabeled 14-kDa polypeptide in their study is Z protein itself.
DISCUSSION
We have purified bovine intestinal and hepatic Z protein using sequential gel filtration, chromatofocusing, and hydroxyapatite chromatography. Both intestinal and hepatic proteins were eluted at similar pH on chromatofocusing and have the same molecular weight. These data as well as similar binding properties for various ligands by both proteins indicate that intestinal Z protein has very similar characteristics to hepatic Z protein. In the rat, three types of fatty acid binding proteins (FABPs) have been isolated and characterized (26, 27). Among these, both intestinal (I-FABP) and liver-type (L-FABP) FABPs have been identified in the rat intestine (26,27). Although FABP has been purified from bovine liver (28), the existence of two types of FABPs in the bovine intestine is not yet known. Since we used ANS binding to monitor the identification of Z proteins in various puri-
O-M
‘-0
,uM
%
Ob
1000,uM
FIG. 7. Inhibition of the fluorescence of ANS (15 pM) bound to bovine hepatic Z protein (0.13 pM) by various ligands. The experimental conditions and abbreviations are the same as those in Fig. 6.
BILE
ACID
BINDING
BY INTESTINAL TABLE
Comparison
of Dissociation
Constants
(PM) for the Binding
Z PROTEIN
155
I
of Bile Acids,
Organic
Anions,
and Fatty
Acids by Z Proteins
Bovine Intestinal Lithocholic acid Taurolithocholic acid Chenodeoxycholic acid Taurochenodeoxycholic acid Deoxycholic acid Taurodeoxycholic acid Ursodeoxycholic acid Cholic acid Taurocholic acid Bilirubin BSP Indocyanine
green
Palmitic acid Stearic acid Oleic acid
Hepatic
0.7 (0.2) 2.8 21 (15) 350 44 350 260
Rat hepatic”
0.5 (0.3) 3.2 28 (33) 70 28 130 140
97 1700
Human hepatic”
4.8 (2.2) 7.0 71 35 6.2 25 32
160 1000
12 (9.4) 14 18 32 5.4 13 120
690
92 110
280
5.2 7.5 2.6
1.4 7.5 5.7
0.27
1.5
1.9
0.8 6.2
1.8 5.2
ND ND
ND ND
0.07
0.04
0.78 0.25
0.69 ND
Note. Dissociation constants were obtained by the ANS displacement method or direct method using equilibrium ’ Data for rat and human hepatic Z proteins are from our previous report (9).
ND
dialysis
(in parentheses).
In conclusion, Z protein may participate in the mechanism of intracellular transfer of bile acids from the brush border to the basolateral pole in enterocytes. Further investigations will be needed to examine the binding properties of the Y’ and Y proteins from the intestinal mucosa.
14. Ketley, J. N., Habig, W. H., and Jacoby, W. B. (1975) J. Biol. Chem.
REFERENCES
17. Stolz, A., Takikawa, H., Sugiyama, Y., Kuhlenkamp, witz, N. (1987) J. Clin. Inuest. 79, 427-434.
1. Wilson, F. A., and Treanor,
L. L. (1981) Gustroenterology
81, 54-
60. 2. Weinberg, S. L., Burckhardt, 3. 4. 5. 6.
G. B., and Wiloson, F. A. (1986) J. Clin. Inuest. 78, 44-50. Lin, M. C., Weinberg, S. L., Kramer, W., Burckhardt, G. B., and Wilson, F. A. (1988) J. Membr. Biol. 106, l-11. Simon, F. R., Sutherland, ,J., and Sutherland, E. (1990) Am. J. Physiol. 259, G394-G401. Wilson, F. A. (1981) Am. J. Physiol. 240, G83-G92. Sugiyama, Y., Yamada, T., and Kaplowitz, N. (1983) J. Biol. Chem.
258,3602-3607. 7. Takikawa, H., Sugiyama, Y., and Kaplowitz, N. (1986) J. Lipid Res. 27,955-966. 8. Takikawa, H., Stolz, A., Sugi.moto, M., Sugiyama, Y., and Kaplowitz, 9. 10. 11.
12. 13.
N. (1986) J. Lipid Res. 27, 652-657. Takikawa, H., and Kaplowitz, N. (1986) Arch. Biochem. Biophys. 251, 385-392. Takikawa, H., and Kaplowitz, N. (1988) J. Lipid Res. 29, 279-286. Takikawa, H., Sugiyama, Y., and Kaplowitz, J. (1988) Biochem. Biophys. Acta 954, 37-43. Kaplowitz, N., Percy-Robb, I. W., and Javitt, N. B. (1973) J. Exp. Med. 138,483-487. Habig, W. H., Pabst, M. J., Fleischner, G., Gatmaitan, Z., Arias, I. M., and Jacoby, W. B. (1974) Proc. Natl. Acad. Sci. USA 71, 3879-3882.
250,8670-8673. 15. Sugiyama, Y., Sugimoto, M., Stolz, A., and Kaplowitz, Biochem. Pharmacol. 33, 3511-3513.
N. (1984)
16. Takikawa, H., Sugiyama, Y., Stolz, A., Sugimoto, M., and Kaplowitz, N. (1986) Biochem. Pharmacol. 35, 354-356. J., and Kaplo-
18. Takikawa, H., Stolz, A., Sugiyama, Y., Yoshida, H., Yamanaka, and Kaplowitz, N. (1990) J. Biol. Chem. 265, 2132-2136. 19. Levi, A. J., Gatmaitan,
M.,
Z., and Arias, I. M. (1969) J. Clin. Znuest.
48,2156-2167. 20. Ockner, R. K., Manning,
J. A., Poppenhausen, W. K. L. (1972) Science 177, 56-58.
21. Takikawa,
H., Stolz, A., and Kaplowitz,
R. B., and Ho,
N. (1987) J. Clin. Inuest.
80,852-860. 22. Takikawa, H., Ookhtens, M., Stolz, A., and Kaplowitz, Clin. Inuest. 80, 861-866.
23. Glantz, J. F. C., Janssen, A. M., Baerwaldt,
N. (1987) J.
C. C. F., and Veerkamp,
J. H. (1983) Anal. Biochem. 837, 57-66. 24. Laemmli,
U. K. (1970) Nature 227, 680-685.
25. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 227, 680-385. 26. Bass, N. M. (1985) Chem. Phys. Lipids 38, 95-114.
27. Bass, N. M. (1988) Znt. Reu. Cytol. 3, 143-184. 28. Haunerland, N., Jagschies, G., Schulenberg, H., and Spener, F. (1984) Hoppe-Seyler’s
2. Physiol. Chem. 365, 365-376.
29. Lowe, J. B., Sacchettini, J. C., Laposata, M., McQuillan, Gordon, J. I. (1987) J. Biol. Chem. 232,5931-5937. 30. Lin, M. C., Kramer, W., and Wilson, 265, 14,986614,995.
J. J., and
F. A. (1990) J. Biol. Chem.
